Insight Into the Heterogeneous Risks of Epidemic Burden After Relaxing Interventions by a Multi-Regional, Age-Stratified Mathematical Model.

In order to capture the regional differences, age stratification, and urban-rural disparities in the dynamics of epidemic risks following eased restrictions, we developed a multilayered modeling framework that integrated social contact patterns with heterogeneous age-demographic structures and incorporated dynamic population mobility networks across diverse regions. Our simulations of the COVID-19 pandemic unveiled spatiotemporal heterogeneity in epidemic trajectories, with the peak, duration, and burden varying significantly across provinces and demographic groups in China. The difference in peak times, up to half a month, provides actionable insights for dynamically allocating medical resources. Regional disparities were further evident, manifested by delayed epidemic peaks in rural areas relative to urban centers. Notably, older populations in rural regions exhibited a disproportionate vulnerability to severe outcomes, highlighting critical health equity concerns. We further investigated how individual behavioral risk preferences, particularly shifts in social contact patterns following the relaxation of restrictions, reshaped epidemic outcomes. Scenario analyses demonstrated that an immediate return to pre-pandemic social interactions could precipitate near-universal infection within 3 months, concurrent with a 2-month shortage of hospital beds. Conversely, isolating infected individuals could enormously mitigate peak infections, particularly protect elderly populations, albeit prolong the overall duration of this epidemic. However, rebound mass gatherings could extend both the duration and complexity of epidemics.